The major focus of this project is to determine whether left hemisphere brain-damaged phonemic/deep dyslexics, who read without the use of letter-sound correspondence rules, and who show a word class dependency (e.g., abstract vs. concrete; nouns vs. function words) in oral reading, are relying on right hemisphere mechanisms for reading. These patients are compared with left hemisphere damaged but non-deep dyslexic patients on reading tasks involving split-field presentation. In addition, evidence for a similar right hemisphere reading capacity is sought in normal readers by examining performance across word categories in split field tasks.